In a recent video titled "Are You A Fit For Usage-Based Pricing?" from the "AI, SaaS & Agentic Pricing with Monetizely" channel, pricing expert Ajit discusses a data-driven approach for SaaS companies considering a transition from user-based pricing to usage-based pricing. Rather than engaging in theoretical debates, he offers tactical guidance on analyzing your existing data to determine whether such a switch would benefit your business.
The Common Usage-Based Pricing Dilemma
Many larger SaaS companies today find themselves contemplating a shift from traditional user-based pricing to usage-based models. On paper, this transition presents multiple options—but which usage metric should you choose? As Ajit points out, internal debates about these decisions can consume significant time and resources.
"Many times, larger companies today are going to consider moving from a user-based pricing model to a usage-based pricing model. And on paper, you may have options, right? You may have options of where should we move to X metric or Y metric? And there's going to be debate inside a company," explains Ajit in the video.
What's often missing from these discussions is a practical analysis of actual customer data to validate whether a proposed pricing change makes mathematical sense.
A Data-Driven Approach to Validation
The key to making this decision successfully lies in examining your existing data. Ajit demonstrates this with a hypothetical contact center software company currently using a per-user pricing model across three tiers (basic, pro, and enterprise).
The company is considering two potential usage metrics:
- Dollars per minute per month (based on call duration)
- Dollars per customer interaction per month (based on number of customers served)
To analyze the viability of these options, you'll need to combine:
- CRM data (customer information, current pricing plans, revenue figures)
- Product usage data (how customers actually use your product)
"I want to be more tactical and show you how to analyze your data set that you may get from your CRM and perhaps someone from the product team that has some usage data and using the usage data and CRM data to figure out whether a particular move is going to be beneficial or not," says Ajit.
What Your Data Analysis Should Include
In the demonstration, Ajit shows how to create a comprehensive data set that includes:
- Customer ID and company name
- Current plan tier
- Number of users per customer
- List prices and discount percentages
- Monthly and annual recurring revenue
- Usage metrics being considered (minutes used, customer interactions)
- Industry and geographic information
This combined data set enables you to perform correlation analyses that are critical in determining which usage metric might make sense as a pricing basis.
The Ideal Scenario: When Usage Correlates With Seats
In the first example, Ajit demonstrates an ideal scenario where usage (total minutes) correlates strongly with the number of users. When plotted on a graph, customers with more users also use more minutes in a relatively predictable pattern.
"What I'm happy to see is that the usage correlates with the number of users and there is an even distribution alongside that. So now if I picked up a dollar per minute use metric, I could more or less mimic the amount of revenue that is coming in today from my existing customers from this new metric," Ajit explains.
This represents a best-case scenario where transitioning to the new usage-based metric would maintain revenue predictability without disrupting customer relationships.
The Reality Check: Uneven Usage Patterns
Ajit then presents a more realistic scenario where usage doesn't correlate well with seat counts. In this second data set, while some customers have many users, they don't necessarily use the product proportionally:
"You'll see a lot of customers pay you a lot of money and have a number of users on the system, but they don't actually make any calls from your system."
This brings up a fundamental question about your product's nature: "Is this a taxi or a car that you keep?" Some products are valued for their availability regardless of frequent use (like a car in your garage), while others are pure consumption-based services (like a taxi).
If your data shows this pattern, switching to minutes used as a pricing metric could significantly disrupt your revenue and potentially alienate high-value customers who don't use the product intensively.
Finding the Right Usage Metric
When the initial metric fails to correlate, you need to explore alternatives. In the example, Ajit shows that while minutes used doesn't correlate well with seats, the number of unique customer interactions does track with the number of users.
"Total number of unique customer interactions become a lot more interesting for me to use as a usage-based metric because even if you're not making calls from my system, you are serving a lot of customers."
This insight demonstrates why examining multiple potential usage metrics is essential—the right one should align with how customers derive value from your product, not just how intensively they use it.
Strategic Considerations Beyond Mathematics
Once you've identified a mathematically viable usage metric, Ajit emphasizes the importance of considering customer psychology and strategic business goals:
"What you then need to figure out is what works better for the customer. Do they want to track how many calls they've made or do they want to track how many customers they've served?"
He suggests that metrics tied to customer outcomes (like customers served) often create less anxiety than those tied to consumption (like minutes used). This distinction is particularly important for enterprise and mid-market customers who need predictable costs.
"As a vendor providing such a service, it might be helpful for you to consider whether a particular metric that you are trying to offer is causing customer anxiety."
Additionally, usage metrics that cause customers to minimize product use to control costs can harm your product adoption goals—a strategic consideration that goes beyond immediate revenue concerns.
Taking Action With Your Own Data
The approach demonstrated in the video can be replicated with your own company data. By combining CRM and product usage information, you can quickly determine whether your proposed usage-based pricing model makes mathematical sense before investing significant resources in a transition.
The key steps include:
- Gather comprehensive customer and usage data
- Analyze current pricing patterns by tier
- Test correlations between seat count and potential usage metrics
- Consider both mathematical viability and strategic implications
This data-driven methodology can save months of theoretical debate and prevent potentially disruptive pricing changes that don't align with how customers actually use and value your product.
Conclusion
Before making the leap to usage-based pricing, analyze your actual customer data to validate whether the proposed metrics correlate with your current revenue model. The right usage metric should not only work mathematically but also align with how customers derive value from your product and support your broader adoption goals.
As Ajit concludes: "Changing to a different metric has a bunch of different considerations. Some of the considerations, what works for customers, some of them are what works for you. And what works for you is also economic, but also strategic."
By taking this pragmatic, data-driven approach to pricing model transitions, you can avoid months of unproductive debate and make decisions based on your actual customer usage patterns rather than theoretical assumptions.